292 



REPORTS ON THE STATE OF SCIENCE, ETC. 



equivalent mortalities in these two broods occur at concentrations differing 

 by some eight units, representing a difference of more than two-fold in the 

 concentration of the poison. The technique adopted of subdividing each 

 brood into six successive strengths, while adequate to cope with small 

 inequalities in resistance between broods, is quite inadequate to yield 

 intelligible results so long as differences of the magnitude illustrated con- 

 tinue to occur. Such differences, moreover, undermine the efficacy of the 

 selective process, by which the parents are chosen, by allowing relatively 

 susceptible individuals from broods in which the survival rate has happened 

 to be high to be mistaken for individuals which have themselves displayed 

 high resistance. 



The stage which the research has now reached is, therefore, primarily 



■'2 ^3 



GENE RATION 



Fig. I. 



one of the improvement of the technique of experimental elimination to a 

 level of efficiency comparable with that already attained in propagation. 

 The stricter control of water temperature and of food density during the 

 test period and of the possibility that some broods would show a considerable 

 mortality even without toxic elimination, are the three methods now being 

 tried to improve this aspect of our procedure. 



Nevertheless, it is obvious from the records that substantial changes in 

 resistance, such as cannot be ascribed to accidental differences, have occurred 

 in at least four out of the five lines, the improvement being often continued 

 progressively from generation to generation. The data from line C6 are 

 shown in Table VIII (at the end of the report) the changes in the estimated 

 concentration for 50 per cent, mortality being given in Table IV, and shown 

 diagrammatically in Fig. i. 



In contrast, no appreciable progress whatever has been achieved in the 

 parallel line C4, for which the estimated 50 per cent, death points are given 

 in Table V. There has been, superficially, a drop in resistance in this line. 



